A central characteristic of alcoholism and other addictive behaviors is the chronically relapsing nature. Despite continuing care's potential for reducing relapse, its implementation is not wide spread due in part to cost, geographic distance, lack of time and peer support. This proposal builds upon the premise that these barriers can be reduced with computer based support systems. We are the national program office for the Network for Improvement of Addiction Treatment (NIATx) and developers of the Comprehensive Health Enhancement Support System (CHESS), a non-commercial eHealth system shown to improve quality of life and produce behavior change. We propose to adapt CHESS for alcohol-dependency relapse prevention and test it in a randomized clinical trial. Addiction CHESS (ACHESS) will be delivered through a new modality (smart-phone) to broaden access, focus on new patients (alcohol dependent patients leaving residential care), offer optional audio delivery to provide access to the functionally illiterate and be enhanced with services tailored to relapse prevention: 1) communication with peer support groups and addiction experts; 2) monitoring of relapse risks, 3) alerts and reminders to encourage adherence to therapeutic goals; 4) addiction-related educational material and tools tailored to individual patients and 5) one-touch communication with a care manager. Research hypotheses include: Primary Outcome. ACHESS + standard care (vs standard care control group) will reduce days of risky drinking over a 12-month period. Secondary Outcomes. ACHESS will reduce proximal outcomes of cravings, negative affect, withdrawal symptoms and temptation events, and produce distal outcomes of increased total abstinence, days of abstinence and reduced negative consequences of substance use. Mechanism of effect. We hypothesize a mediational model where ACHESS improves competence, relatedness and autonomy, which will improve outcomes. We will recruit a final sample of 280 alcohol dependent patients (30% minorities) from two NIATx addiction treatment agencies. Data sources will be interviews at pretest, 3, 6, 9 and 12 months, chart reviews, lab tests and automatic collection of computer use data. This is a resubmission of 1R01 AA01792-01.